Classification of ECG Using Ensemble of Residual CNNs with Attention Mechanism

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68081731%3A_____%2F21%3A00555163" target="_blank" >RIV/68081731:_____/21:00555163 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/document/9662723" target="_blank" >https://ieeexplore.ieee.org/document/9662723</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.23919/CinC53138.2021.9662723" target="_blank" >10.23919/CinC53138.2021.9662723</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Classification of ECG Using Ensemble of Residual CNNs with Attention Mechanism

Popis výsledku v původním jazyce

This paper introduces a winning solution (team ISIBrno-AIMT) to the PhysioNet Challenge 2021. The method is based on the ResNet deep neural network architecture with a multi-head attention mechanism for ECG classification into 26 independent groups. The model is optimized using a mixture of loss functions, i.e., binary cross-entropy, custom challenge score loss function, and sparsity loss function. Probability thresholds for each classification class are estimated using the evolutionary optimization method. The final model consists of three submodels forming a majority voting classification ensemble. The proposed method classifies ECGs with a variable number of leads, e.g., 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead. The algorithm was validated and tested on the external hidden datasets (CPSC, G12EC, undisclosed set, UMich), achieving a challenge score 0.58 for all tested lead configurations. The total training time was approximately 27 hours, i.e., 9 hours per model. The presented solution was ranked first across all 39 teams in all categories.

Název v anglickém jazyce

Classification of ECG Using Ensemble of Residual CNNs with Attention Mechanism

Popis výsledku anglicky

This paper introduces a winning solution (team ISIBrno-AIMT) to the PhysioNet Challenge 2021. The method is based on the ResNet deep neural network architecture with a multi-head attention mechanism for ECG classification into 26 independent groups. The model is optimized using a mixture of loss functions, i.e., binary cross-entropy, custom challenge score loss function, and sparsity loss function. Probability thresholds for each classification class are estimated using the evolutionary optimization method. The final model consists of three submodels forming a majority voting classification ensemble. The proposed method classifies ECGs with a variable number of leads, e.g., 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead. The algorithm was validated and tested on the external hidden datasets (CPSC, G12EC, undisclosed set, UMich), achieving a challenge score 0.58 for all tested lead configurations. The total training time was approximately 27 hours, i.e., 9 hours per model. The presented solution was ranked first across all 39 teams in all categories.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20601 - Medical engineering

Projekt

<a href="/cs/project/FW01010305" target="_blank" >FW01010305: Umělá inteligence pro autonomní klasifikaci EKG v rámci online telemedicínské platformy</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

2021 Computing in Cardiology (CinC)

ISBN

978-166547916-5

ISSN

2325-8861

e-ISSN

2325-887X

Počet stran výsledku

4

Strana od-do

14

Název nakladatele

IEEE

Místo vydání

New York

Místo konání akce

Brno

Datum konání akce

12. 9. 2021

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—